LIGHTNING PROTECTION OF THATCHED ROOFED
STRUCTURES
FERNANDO HAUSSE CHACHAIA
A research report submitted to the Faculty of Engineering and the Built Environment,
University of the Witwatersrand, in partial fulfilment of the requirements for the degree
of Master of Science in engineering.
Johannesburg 2006
Declaration
I declare that this Research Report is my own, unaided work, except where
otherwise acknowledged. It is being submitted for the degree of Master of Science in
Engineering in the University of the Witwatersrand, Johannesburg. It has not
been submitted before for any degree or examination in any other university.
Signed this ____ day of ___________ 20___
_________________________________
Fernando Hausse Chachaia
i
Dedication
I dedicate this work to my daughter and my two sons Evelina, Sandro and Fernando Jr.,
who are always in my mind. I also dedicate it to my wife Paulina for her courage and
patience through all this time, and finally I dedicate it to the memory of my father and
mother, Hausse Chachaia and Evelina Joao, who in life taught me throughout my
childhood how to keep a sense of pragmatism and perseverance in every issue.
ii
Acknowledgements
I would like to express my most sincere gratitude to my supervisor, Prof I.R. Jandrell, for
his support over the course of this research report.
I would like to thank SANTED PROJECT for giving me finance support during my
studies.
Finally I would like to thank my friends and colleagues, especially Mr Andrew Dickson
and Mr Nicholas West for useful comments and for their emotional support.
iii
Table of Contents
Abstract.............................................................................................................................. 1
1
Introduction............................................................................................................... 2
1.1
Research Report Overview ................................................................................. 3
2
Background ............................................................................................................... 5
3
Nature of lightning.................................................................................................... 6
3.1
Formation of lightning ........................................................................................ 6
3.2
Cloud-to-ground flashes...................................................................................... 7
3.2.1
Discharge process ....................................................................................... 7
3.2.2
Negative flash to the ground ....................................................................... 7
3.2.3
Positive flash to the ground......................................................................... 8
3.2.4
Final stage of lightning ............................................................................... 9
3.3
4
3.3.1
The peak lightning current ........................................................................ 10
3.3.2
The rate of rise of the lightning current .................................................... 11
3.3.3
Lightning current waveform ..................................................................... 12
3.3.4
Subsequent strokes.................................................................................... 14
Effects of lightning .................................................................................................. 15
4.1
5
Direct effects..................................................................................................... 15
4.1.1
Thermal effects ......................................................................................... 15
4.1.2
Mechanical effects .................................................................................... 16
4.1.3
Sparking effects ........................................................................................ 16
Assessment of risk of damage due to lightning .................................................... 17
5.1
6
Lightning currents and related parameters........................................................ 10
Risk calculation and determination of protection level .................................... 17
5.1.1
Calculation of the equivalent collection area, Ae ...................................... 18
5.1.2
Calculation of LPS efficiency, E .............................................................. 19
5.1.3
Expected number of direct strike .............................................................. 21
5.1.4
Estimated efficiencies of an LPS and its related levels of protection....... 22
Methods of lightning protection ............................................................................ 24
6.1
The protective angle method............................................................................. 24
iv
7
8
9
6.2
The rolling sphere method ................................................................................ 26
6.3
The mesh method.............................................................................................. 28
Components of lightning protection system (LPS) .............................................. 30
7.1
Air terminations ................................................................................................ 30
7.2
Down conductors .............................................................................................. 31
7.3
Bonding conductors .......................................................................................... 32
7.4
Earth termination .............................................................................................. 33
Lightning protection system for thatched roofs................................................... 35
8.1
Lightning mast .................................................................................................. 35
8.2
Lightning conductors above the roof ................................................................ 37
8.3
Protection of equipment inside the thatched roofed structure .......................... 37
High voltage laboratory tests ................................................................................. 39
9.1
Test description................................................................................................. 39
9.2
Discussions of results........................................................................................ 40
10
Conclusions and recommendations ....................................................................... 45
11
List of references ..................................................................................................... 47
12
Appendices............................................................................................................... 48
12.1
Lightning ground flash density, Ng ................................................................... 48
12.2
Determination of the environmental coefficient, Ce ......................................... 49
12.3
Calculation of the accepted annual frequency of lightning flashes, N c ........... 49
12.4
Test and measurement system .......................................................................... 52
12.5
Equivalent circuit .............................................................................................. 53
v
List of Figures
Figure 3-1:
Charge in typical cumulonimbus cloud ...................................................... 6
Figure 3-2:
Negative downward leaders........................................................................ 7
Figure 3-3:
Positive downward leaders ......................................................................... 8
Figure 3-4:
Downward and upward leaders................................................................... 9
Figure 3-5:
Experimental statistical distribution of positive and negative lightning
strokes as a function of their amplitude [2] .............................................................. 11
Figure 3-6:
Experimental statistical distribution of positive and negative lightning
strokes as a function of front steepness [2]............................................................... 12
Figure 3-7:
Double exponential current waveform...................................................... 13
Figure 3-8:
Lightning flash current waveforms ........................................................... 14
Figure 5-1:
The equivalent collection area of a structure ........................................... 19
Figure 6-1:
An imaginary cone formed from protective angle [1] .............................. 26
Figure 6-2:
Rolling sphere [11] ................................................................................... 27
Figure 6-3:
Space protected by two air terminations [1] ............................................. 28
Figure 8-1:
Lightning protection system using one mast ............................................ 36
Figure 8-2:
Lightning protection system using two masts........................................... 36
Figure 9-2:
Impulse current waveforms from measurements with 20 mm of airgap .. 43
Figure 9-3:
Impulse current waveforms from measurement with 30 mm of airgap.... 44
Figure 13-1:
Test and measurement system .................................................................. 52
Figure 13-2:
Equivalent circuit of current impulse generator (8/20 µs) [13]................. 53
Figure 13-3:
Equivalent circuit of combined (Voltage, Current) impulse generator
(1.2/50 µs, 8/20 µs) [13]............................................................................................ 53
vi
List of Tables
Table 3-1:
The experimental statistical distribution of lightning strikes as a function
of their amplitude ...................................................................................................... 10
Table 3-2:
The lightning values.................................................................................. 13
Table 5-1:
The efficiency, E and its related protection level ..................................... 20
Table 6-1:
Protective angle according to the protection level .................................... 25
Table 6-2:
Rolling sphere radius ................................................................................ 26
Table 6-3:
Recommended size for mesh .................................................................... 29
Table 7-1:
Average distance between down conductors ............................................ 31
Table 7-2:
Minimum dimensions for bonding conductors ......................................... 32
Table 7-3:
Soil resistivity and corrosiveness.............................................................. 34
Table 9-1:
Standard exponential impulse current according to SABS IEC 60-1 ....... 39
Table 9-2:
Parameters from test with 10 mm of airgap.............................................. 41
Table 9-3:
Parameters from test with 20 mm of airgap.............................................. 42
Table 9-4:
Parameters from test with 30 mm of airgap.............................................. 43
Table 13-1:
Lightning ground flash density Ng in South Africa [1]............................. 48
Table 13-2:
Environmental coefficient, Ce [1] ............................................................. 49
Table 13-3:
Factors depending on the type of structure [1] ......................................... 49
Table 13-4:
Factors depending on the contents of structure [1]................................... 50
Table 13-5:
Factors depending on the consequential losses [1] ................................... 51
vii
Abstract
This work describes the design of external lightning protection system for thatched
roofed structures. Lightning is a natural phenomenon that has an unpredictable strike
location. Therefore in creating an external LPS, it is possible to create a path for the
discharge that minimizes its harmful effects to a structure. There are many international
documents and standards that describe methods of providing lightning protection. It has
traditionally been accepted practice to install lightning protection masts near a thatched
structure. However in solving one problem a number of others may have been created.
This work explores alternative lightning protection designs for thatched roofed structures.
Tests were performed in a laboratory environment to determine the level at which that
will ignite from a direct strike to thatch material. These tests were performed using a
combination impulse generator. It was found that thatch started to smoke at 10 kA. And
ignition of the thatch occurred at 20 kA with a charge of 45 C.
1
1
Introduction
Lightning has higher destructive effect when the structure is not protected. In order to
minimize damage in investments, protection against lightning is required and extends life
of structures. Therefore protection is needed to avoid deaths of people who are living and
working in thatched roofed structures in the instant that lightning hits the structure.
The main aim of this research is to design a system for protecting thatched roofed
structures. No device can stop lightning formation. However, it is possible to create a
path for the discharge that minimizes its harmful effects on the structure.
The design and installation of LPS must comply with IEC, BS and other relevant
standards for protection against lightning if it is to be effective.
The lightning protection system established from these standards must shield a thatched
roofed building, its occupants and equipment from the adverse effects associated with a
lightning strike. These effects could otherwise result in fire, structural damage and
electrical interference. To perform correctly, the protection system must capture the
lightning, lead it safely downwards and then disperse the energy in the ground.
2
1.1 Research Report Overview
This research report is organized in the following way:
Chapter 1 and 2 provide the introduction and background to the topic.
Chapter 3 covers the theoretical analysis of the nature of lightning. The formation,
mechanism and parameters of lightning are described in this chapter.
Chapter 4 presents the main direct effects of lightning. on thatched roofed structures.
Chapter 5 covers the assessment of the risks of the damage caused by lightning.
Principles and equations to calculate the risk of damage due to lightning are given in this
chapter.
Chapter 6 presents a review of different methods of lightning protection. The protective
angle method, the rolling sphere method and the mesh method are discussed in this
chapter.
Chapter 7 presents the components of the lightning protection system. Components such
as air termination, down conductors, bonding conductors, earth-termination and surge
protection devices are described.
Chapter 8 covers the lightning protection systems for thatched roofed structures. For
lightning protection systems, one method is commonly used for protecting thatched
roofed structures, - the mast method. There are however other methods. In this chapter
we look at two methods, the mast and conductors above the roof.
Chapter 9 covers the high voltage test. The description and the results of the test are
presented in this chapter.
3
Chapter 10 includes the overall conclusions of this research report and the
recommendations for additional measures for protecting thatched roofs.
4
2
Background
In South Africa most of the rural communities’ houses and a whole lot of up-market
developments are thatched. A thatched roof is constructed with soft material, such as
straw, reed, grass or coconut leaves. This material is highly flammable. According to the
South African Code thatch is particularly prone to ignite because it is liable to become
fluffy at surface and, if moist [1], methane and other flammable gases can be formed.
Particular attention must be paid to the protection of thatched roofs. South African
Standard and other Standards, such as the British standard provide guide for the lightning
protection of thatched roofed dwelling.
The design of lightning protection for thatched roofed dwellings must be considered in
order to provide paths for dangerous lightning currents to the ground, to avoid sparking
as far as possible and to avoid secondary direct contact between the hot lightning channel
and the roof cover.
The protection of thatched roofed structures against lightning is provided by lightning
masts. They do this by providing a preferential point of strike and conductive paths for
lightning currents to follow. This diverts the currents away from the structure and not
through it.
To review the engineering models involved in the conventional approach to lightning
protection, the attachment of the leader to the struck object is often described using the
so-called electro-geometrical theory [2], the core of which is the concept of a “striking
distance”. This concept obscures some of significant physics but allows the development
of relatively simple and useful techniques for designing conventional lightning protection
systems [2].
5
3
Nature of lightning
To provide a good lightning protection system for thatched roofed structure it is
important to understand the nature of lightning.
3.1 Formation of lightning
It is generally accepted that lightning is created by a separation of electric charges due to
air turbulence, although they can occur in other atmospheric conditions. The charges
separation is thought to be due to the complex processes of freezing and melting and by
movements of raindrops, snowflakes and ice crystals involving collisions and splintering.
Clouds containing moisture rise and cool as they do so [2, 3]. If the rate of the rise is
gradual, then mist and rainfall normally result. However, if the rate of the rise is above a
certain level, the cooling effect will be accelerated. This may cause larger raindrops or
even freezing. Typically, most positive charges accumulate at the top of the
cumulonimbus clouds, leaving the lower regions negative, although there may be a small
positive region near the base, as shown in Figure 3-1 [2, 3].
Figure 3-1:
Charge in typical cumulonimbus cloud
6
3.2 Cloud-to-ground flashes
3.2.1 Discharge process
For cloud-to-cloud lightning the initiating and terminating charge regions are both in a
thundercloud and can cause electrical interference and sometimes significant damage,
while for cloud-to-ground lightning the initiating charge region is in a thundercloud and
the terminating charge region is on the ground and it is the ground strikes which are
generally the most destructive [3, 4]. As the potential difference between the cloud base
and the underlying air/ground plane exceeds the breakdown value of the air in the
immediate vicinity, the air becomes ionized and a down-stroke begins travelling at about
2 meters per micro-second. It follows a haphazard path, generally downwards, made up
of small steps [2, 3].
3.2.2 Negative flash to the ground
There is some debate about the way in which the steps are produced and the point at
which the actual arc commences, but eventually the negatively charged downwards
leader will approach the ground as shown in Figure 3.2 [3, 5].
Figure 3-2:
Negative downward leaders
7
3.2.3 Positive flash to the ground
Positive flashes to the ground pose a special problem since they do not strike the highest
point (or any air termination system provided for the purpose). They generally occur less
frequently than negative flashes, however in certain geographic locations there may be
more positive flashes to ground. Present standards have assumed an average of around
10% positive flashes to ground [3, 5]. Normally they consist of one stroke only. They
have slower rise times than negative flashes, with high peak current and charge transfer;
the duration is longer than a single stroke of a negative flash but usually shorter than a
complete negative flash [3, 5]. The stroke may be followed by a continuous current. The
Figure 3.3 shows the positive flash to ground [2, 5].
Figure 3-3:
Positive downward leaders
8
3.2.4 Final stage of lightning
Positive charges will, in turn, be induced on the ground surface and, in particular on any
high structures. If the potential is sufficiently high at the ground (or on a high structure),
then ionisation of the air commences and an upward positively charged leader, will be
created as shown in Figure 3-4 [5]. Eventually the negative and positive charged leaders
will meet, often via a seemingly haphazard route, and the lightning discharge will be
experienced, which is normally negative.
Figure 3-4:
Downward and upward leaders
The amount of lightning activity is not the same in South Africa ranging from the
extreme (Mpumalamga) to the nominal (Southem Cape), it varies according to many
factors, including geographical location, height etc [5]. The energy associated with the
discharge also varies. It is necessary to consider these factors and others, in deciding
whether a lightning protection system is required and the form it should take.
9
3.3 Lightning currents and related parameters
When an object on the ground, be this a thatched roofed structure or any other structure,
is struck by lightning the stresses to which it is subjected are determined by the current
discharged into it and is best described in terms of associated current waveform. From the
point of view of lightning protection this current represents the most important single
parameter of the lightning discharge. The measurements of the stroke currents on the
ground have shown that the high current is characterized by a fast rise to crest (1 to 10
µs) followed by a longer decay time of 50 to 1000 µs to half-time [2, 5].
3.3.1 The peak lightning current
The peak current related to the lightning stroke has amplitudes between 2 kA and 200 kA,
(1 % of strokes exceed 200 kA) and are assumed statistical lognormal distributions as
shown in Table 3-1 [2] and Figure 3-6 [2]. The time that the lightning stroke reaches the
peak value is about 10 µs [2, 5].
Table 3-1:
The experimental statistical distribution of lightning strikes as a
function of their amplitude
Percentage of lightning stroke
The peak lightning current
[kA]
1
Of the strokes exceed
200
5
Of the strokes exceed
100
10
Of the strokes exceed
80
20
Of the strokes exceed
50
50
Of the strokes exceed
28
60
Of the strokes exceed
20
90
Of the strokes exceed
8
99
Of the strokes exceed
3
10
Figure 3-5: Experimental statistical distribution of positive and negative lightning
strokes as a function of their amplitude [2]
3.3.2 The rate of rise of the lightning current
The rate of rise of the current,
di
, is important when calculating voltdrops across
dt
inductive components of lightning protection systems, and for calculating induced
voltages due to adjacent lightning strikes.
The probabilities of the maximum rate of rise of the current of a negative lightning stroke
and positive lightning stroke are illustrated in Figure 3-7 [2].
11
Figure 3-6: Experimental statistical distribution of positive and negative lightning
strokes as a function of front steepness [2]
3.3.3 Lightning current waveform
The standard lightning is comprised of current waveform components which present the
important characteristics of natural lightning flashes and is approximated by a double
exponential waveform. The waveform parameters defined in SABS IEC 61024-1-1
(1993) are shown in Figure 3-8 [2, 5]. The lightning current waveform is described in
terms of the time to crest, tcr, and the time to half value, td, so that a 8/20 µs waveform has
a time to crest of 8 µs and time to half value of 20 µs.
12
Figure 3-7:
Double exponential current waveform
These values in Table 3-2 [2] should be compared with the standard waveforms used for
testing electrical equipment.
Table 3-2:
The lightning values
Current
2000 to 200,000 A
Voltage
100 to 1,000,000 kV
Duration
70 to 200 µs
Electrical discharge in the cloud
20 to 50 C
Energy
4 to 10 kWh
Time to crest (voltage)
1.2 µs
Time to half value (voltage)
50 µs
Time to crest (current)
8 µs
Time to half value (current)
20 µs
5.5 kA/ µs
di (t )
dt
13
3.3.4 Subsequent strokes
The subsequent strokes in the flash tend to have a higher rate of rise of current but lower
peak amplitudes than the initial stroke and they can therefore be significant for induced
voltages in wiring, where the inductively coupled voltages are proportional to the rate of
change of the lightning current [2, 5].
The first stroke within a flash usually has the largest peak current (typically three times
that of subsequent strokes), therefore when describing the peak current of a flash we
usually imply that of the first stroke [2, 5].
Near the end of some of the strokes in a negative flash, there is often a lower level current
of a few kA persisting for several milliseconds as illustrated in Figure 3-9 [5].
Figure 3-8:
Lightning flash current waveforms
14
4
Effects of lightning
The effects of lightning can result in fire damage for thatched roofed structures if the
protective measures are not taking into account. To provide these protective measures it
is important to analyse these effects. This chapter discusses the main effects of direct
lightning strike on structures.
4.1 Direct effects
These are caused by current transfer via direct attachment. They will be considered
individually [2, 6]:
4.1.1 Thermal effects
When a lightning current pulse flows through a conductor of resistance R, heat is
generated. Practically the whole of this heat is devoted to raising the temperature, since
no significant portion of the heat can flow to the structure during the very short duration
of the pulse [6].
In the construction process of thatch, wire mesh is used to reinforce and secure the
bundles of thatch. If lightning current flows through the wire mesh, the thermal effect
may occur and it can cause a fire for thatch. To minimize the thermal effects the metalcoated insulating sheets are used in the preparation of thatch [1, 6].
The conductor which may carry the lightning current therefore need to be designed with a
cross-sectional area large enough to keep the temperature rise well below a critical value
such as the ignition point or melting point of the material. The design also needs to
account for the fact that rapidly changing currents in the lightning pulse tend to
concentrate at the surface of the conductor (skin effect) [6].
15
4.1.2 Mechanical effects
The conductors which form a LPS for thatched roof should have a cross-sectional area
large enough so that can carry lightning current without melting or fusing explosively. If
the conductor is subjected to melt especially in contact with thatch, the thatch can ignite
[6].
4.1.3 Sparking effects
The most dangerous effects for thatch are sparking effects. These effects in the presence
of thatch can cause fire risk. Voltage and thermal sparking may occur either separately or
together. Voltage sparking is the result of dielectric breakdown including tracking or
flashover across dielectric surfaces. It could arise inductively in a loop, bend or from the
resistive drop in a high resistance material, especially at joints [2, 6].
Thermal sparking consists of burning fragments of melted material thrown out from hot
spots such as high resistance contacts having a high current concentration, or at acute
changes of geometry. The temperatures of both types of spark are high and are potential
sources of fire or explosion [2, 6].
16
5
Assessment of risk of damage due to lightning
The procedure for calculating the risk is well described in the standards, such as SABS
0313 (1999), SABS IEC 61662-1 (1995) and IEC 62305-2, Ed.1. The result of this
calculation determines the need of a lightning protection system and its protection level.
The risk assessment compares the expected lightning strike with the probability of
lightning strike on the structure. The rate of these two factors indicates the need of a
lightning protection system and its degree of security (protection level) [7, 8]. This value
depends on several factors, such as the type of structure and its contents, although
sometimes other considerations could be taken into account in order to improve the
protection level over the risk calculations [7].
Standards of practice for lightning protection consider that lightning protection is needed
in the following cases [7, 8]:
•
Large concentrations of persons
•
Need of continuity in production or public services
•
Areas with high lightning density
•
Very high or isolated buildings
•
Buildings containing explosive or inflammable materials
•
Buildings containing irreplaceable heritage
•
Buildings or structures whose risk index, calculated according to the standard,
determine the need for a lightning protection system with a certain protection level
5.1 Risk calculation and determination of protection level
The SABS 0313 provides an equation to determine the expected annual frequency of
direct lightning flashes to a structure, N d . The following equation is used:
17
N d = N g × Ae × C e × 10 −6
(5-1)
N g is the average ground lightning flash density in square kilometre per year, concerning
the particular area where the structure is located. This value can be obtained statistically
from an isokeraunic map. In South Africa, the value of N g is given from Table 13-1 in
Appendices. C e is the environmental coefficient which describes the particular location
of the structure and is given in Table 13-2 Appendices.
5.1.1 Calculation of the equivalent collection area, Ae
The equivalent collection area Ae is given in Figure 5-1 and is calculated as follows [1,
5]:
Ae = LW + 6 H ( L + W ) + 9πH 2
(5-2)
The equivalent collection area, Ae, of the structure in square kilometres is defined as the
measure of the ground surface and a straight line with a 1:3 slope which passes from the
upper parts of the structure as shown in Figure 5-1 [1, 5]. This area has the same expected
annual frequency of direct lightning flashes, N d as the structure.
18
Figure 5-1:
The equivalent collection area of a structure
5.1.2 Calculation of LPS efficiency, E
The estimation of LPS efficiency is achieved after analysing the risk of damage. This
analysis takes into account appropriate factors such as [1, 9]:
•
Type of construction;
•
Presence of flammable and exposure substances;
•
Measures put in place to reduce the consequential effects of lightning;
•
Number of people affected by the damage;
•
Type and importance of service concerned; and
•
Value of goods that could suffer damage
19
SABS 0313 also provide an equation to determine the accepted annual frequency of
lightning flashes, N c , as follows:
Nc = a × b × c
(5-3)
a is the Factor that depends on the type of the structure, b is the Factor which depends
on the contents of the structure, c is the Factor which takes into account the
consequential losses. The factors a , b and c are determined from Tables 13-3, 13-4, 135 respectively in Appendices
With these values, the efficiency, E and the protection levels are calculated as follows
[1]:
E = 1−
Nc
Nd
(5-4)
Table 5-1:
The efficiency, E and its related protection level
Calculated efficiency, E
Corresponding protection level
E > 0,98
Level I + additional measures
0,95 < E ≤ 0,98
Level I
0.90 < E ≤ 0.95
Level II
0.80 < E ≤ 0.90
Level III
0 < E ≤ 0.80
Level IV
When the calculated efficiency of the LPS, E is equal to zero no protection measure is
needed [8, 9].
20
The need for lightning protection system is determined from the comparison between the
value of the accepted annual frequency of direct lightning flashes ( N c ) and the expected
annual frequency of direct lightning flashes to the structure ( N d ) as follows [9]:
If N d is less than or equal to N c , the LPS is not needed.
If N d exceeds N c , an LPS of efficiency, E calculated in equation 5-4 can be installed
and the proper protection level can then be selected according to Table 5-1 [5, 9].
5.1.3 Expected number of direct strike
As described in section 5.1, the relationship between expected annual frequency of direct
lightning flashes to a structure ( N d ) and lightning ground flash density ( N g ) is defined
by the equation 5-1. Table 13-1 in Appendices gives the lightning ground flash density in
South Africa.
Considering the cone of protection (Figure 8-1 and Figure 8-2), we can estimate the
effective area of a structure projected over the ground surface, the area of attraction
increases as the height of the mast increases. For a mast height H, a circle of radius 3H
produces a protection cone for the structure that will intercept a lightning flash. Thus,
assuming a thatched roofed building of 10m x 6m, with a height of 9m, the effective area
is shown in Figure 5-1. Using equation 5-2, the calculated area is equal to 3200 m2 =
0.0032 km2.
Applying the lightning ground flash density of Table 13-1 in Appendices for
Johannesburg, we obtain an estimated interception rate for the following different cases
according to the environmental coefficient given on Table 13-2 in appendices as follows:
1) A structure amongst other buildings and trees of the same or greater height
N d = 7.5 × 0.0032 × 0.25 = 0.006 Flashes per year
21
2) A structure surrounded by smaller buildings
N d = 7.5 × 0.0032 × 0.50 = 0.012 Flashes per year
3) Isolated structure or no buildings within a distance of 3H
N d = 7.5 × 0.0032 × 1.0 = 0.024 Flashes per year
4) Isolated structure on a hill
N d = 7.5 × 0.0032 × 2.0 = 0.048 Flashes per year
For each of the above cases the calculated flashes per year are related to the area covered
by the assumed building and its associated cones. The building will be expected to
intercept a lightning flash once every hundred and sixty seven years for case 1 once every
eighty three years for case 2 once every forty two years for case 3 and once every twenty
one years for case 4.
The lightning flash interception rate of the building can be understood as the flash
interception rate of the installed LPS for the structure, since the LPS is designed for
protecting the structure.
5.1.4 Estimated efficiencies of an LPS and its related levels of
protection
From equation 5-3 and Tables 13-3, 13-4, 13-5 in Appendices, the estimated and
accepted annual frequency of direct lightning strikes ( N c ), is equal to 10 −3 Flashes per
year. Therefore, the estimated efficiencies of an LPS and its related levels of protection
for each case are the following:
1)
83% (0.80 < E ≤ 0.90)
-Level III
2)
92% (0.90 < E ≤ 0.95)
-Level II
3)
96% (0.95 < E ≤ 0.98)
-Level I
22
4)
98% (0.95 < E ≤ 0.98)
-Level I
The obtained levels of protection justify the need of LPS in order to protect the thatched
roofed structures, its occupants and equipment from lightning flashes.
Taking into account that thatched roofed structures are prone to ignite, the cases 2, 3 and
4 above are considered in the designing of lightning protection system in this work,
implying that protection level of I and II is necessary for thatched roofed structure.
23
6
Methods of lightning protection
The methods of lightning protection are well described using the zone of protection
concepts for a structure. The zone of protection is the space within the volume to be
protected and where the lightning protection air termination is located. Air termination
can have different forms such as [1, 5]:
•
Rods;
•
Stretched wires; and
•
Meshed conductors.
In order to place the air termination in a proper position on the building to be protected,
the following methods can be used [1, 5]:
•
The protective angle method;
•
The rolling sphere method; and
•
The mesh method
6.1 The protective angle method
The structure to be protected lies within an imaginary cone ABC with the highest point of
the structure providing protection as shown in Figure 6-1 [1]. Table 6-3 [1] provides the
recommended values for the cone angle, α (in degrees).
24
Table 6-1:
Protective angle according to the protection level
Protection
Height of air termination, ht
level
[m]
20
30
45
60
Angle, α [in degrees]
I
25
-
-
-
II
35
25
-
-
III
45
35
25
-
IV
55
45
35
25
The protection angle method is suitable for simple buildings. If the height is larger than
the rolling sphere radius, R, the protection angle method can not be used [10, 11].
The zone of protection offered by an air termination system is considered to be 45º for
heights up to 20m [11]. Above this height, the zone of protection is determined by the
rolling sphere method.
25
Figure 6-1:
An imaginary cone formed from protective angle [1]
6.2 The rolling sphere method
The rolling sphere method is used when we are dealing with complex structures. When
the use of the protective angle method is excluded, the rolling sphere method is used to
identify the protected volume of the structure [1, 11].
This involves rolling an imaginary sphere of radius given in Table 6-2 [10] over a
structure. The areas touched by the sphere are deemed to require protection as shown in
Figure 6-2 [11]. On tall structures, this can obviously include the sides of the building.
Table 6-2:
Rolling sphere radius
Protection level
Radius of the sphere
I
20
II
30
III
45
IV
60
26
Figure 6-2:
Rolling sphere [11]
If two parallel horizontal LPS air terminations are placed above the horizontal reference
plane as shown in Figure 6-3 [1], the penetration distance p of the rolling sphere below
the level of the air terminations is calculated as follows [1]:
d 
p = R− R − 
2
2
2
(6-1)
p is the penetration distance of rolling sphere, R is the radius of rolling sphere and d is
the distance separating two air terminations.
27
Figure 6-3:
Space protected by two air terminations [1]
1- Two air – terminations, 2- horizontal reference plane, 3- whole protected area, ht physical height of the air-termination above the reference plane. This height is given in
Table 6-1.
6.3 The mesh method
The mesh method is suitable for protecting flat surfaces. BS 6651 recommends that on
high risk structures such as explosive factories and thatched roofs, no part of the roof
should be more than 2.5m from an air termination conductor. This is generally achieved
by applying a 5m x 10m mesh to the roof [11].
However, for most structures, a mesh of 10m x 20m is considered sufficient, giving a
maximum distance from any part of the roof to the nearest conductor of 5m. All metal
structures on the roof would be interconnected using this mesh [11].
28
The recommended sizes for the mesh according to their protection level are listed in
Table 6-3 [5, 10].
Table 6-3:
Recommended size for mesh
Protection level
Mesh size [m]
I
5
II
10
III
10
IV
20
29
7
Components of lightning protection system (LPS)
In the design of LPS for thatched roofed structures is based on the concepts of a
protective angle and rolling sphere methods, which are applied to the structure to ensure
that all exposed areas are protected by the system. The lightning protection system must
be designed to provide sufficiently low impedance so that the lightning energy will
follow the required route. This requires an integrated design and use of materials with
sufficiently low impedance. The various components of the lightning protection system
for thatched roofed structures are as follows [5, 11]:
•
Air terminations
•
Down conductors
•
Bonding conductors
•
Earth termination
7.1 Air terminations
In general air terminations consist of a vertical rods, stretched wires and meshed
conductors on the roof and top edges of a structure. The conductors typically form a mesh
of 10 m by 20 m, smaller on high risk buildings. Metal projections, including rods, are
connected to this [11].
For thatched roofed structures, air terminations consist of the highest point of the mast,
stretched wires and meshed conductors on the roof and top edges of a structure.
BS 6651 requires that all parts of the roof are within 5m of a conductor of the air
termination. This distance is reduced to 2.5 m in high risk buildings. Rods, if used, are
positioned near those locations where a strike is most likely (for example roof peaks,
building corners etc).
30
7.2 Down conductors
For thatched roofed structures down conductors are the masts and are required to provide
a low impedance path down the structure, in a manner which minimises potential
differences and induced current.
The mast must be located at least one metre a way from the thatched roofed structure.
This will avoid the flash-side that can occur between the mast and metal objects in the
structure [5].
In general the down conductors would be symmetrically positioned around the building
in such a way that the average value of the distance between them is more than the values
indicated in Table 7-1 [10], ideally including the corners. Sensitive electronic equipment
should not be positioned within the building, near to these down paths as there is a risk of
inductive interference. Currents would flow in all paths, but the largest would flow in the
path nearest to the point of impact.
Table 7-1:
Average distance between down conductors
Protection level
Average distance [m]
I
10
II
15
III
20
IV
25
Each down conductor should be taken to an earth termination and interconnected via a
horizontal ring conductor installed near ground level. A test clamp is often fitted to
enable the continuity of pairs of down conductors to be checked at ground level and
provide a means of isolating the earth electrode [12].
31
7.3 Bonding conductors
In the case of large thatched roofed structures and thatched roofed structures involving a
group of closely spaced houses, the lightning protection system requires more than one
mast. In this case bonding conductors are necessary for interconnecting the masts.
The metalwork (such as central heating ducts, pipes etc.) in the thatched roofed structures
must be connected to the bonding bar using bonding conductors. Bonding bars are
constructed and installed in a way that allows easy access for inspection [1].
The bonding bar should be connected to the earth termination system. For large
structures, more than one bonding bar could be provided and interconnected [12]. This is
to ensure that a side-flash does not occur. As the current flows into conductors, a
potential would be created [13]. If metalwork is not bonded, this can initially be at a
potential nearer to that of earth and thus can offer a more desirable path to ground [13].
The bonding bars should be placed above ground at vertical intervals not exceeding 20 m
for structures of more than 20 m in height [9, 12]. Bonding bars should be connected to
the horizontal ring which bonds the down conductors.
If the whole lightning current or substantial part of it flows through a bonding
connection, the minimum dimension for cross-sections of bonding conductors are given
in Table 7-2 [10].
Table 7-2:
Minimum dimensions for bonding conductors
Protection level
Material
Cross section
[mm2]
I to IV
32
Cu
16
Al
25
Fe
50
7.4 Earth termination
The mast protecting thatched roofed structure can be used at below ground as earth
termination. For thatched roofed structures involving a group of closely spaced houses,
the earth termination can consist of a ring of buried copper which encircles the structure.
As we know from the section 7.2 the down conductors for thatched roofed structures are
the masts, Each mast must have separate earth electrode termination and these are
normally looped together to form a ring, with horizontal electrodes being used The
impedance of the termination is required to be a maximum of 10 Ω [11]. The most
common terminations are driven rods and these are required to be at least 1.5 m long with
a minimum of 9m being used for each system [11].
The ring helps to provide potential equalisation at ground level, in addition to potential
grounding. The latter helps reduce the touch voltage which could be experienced by a
person in contact with the mast during a lightning discharge. The external ring earth
electrode should be used preferably buried at a depth of at least 0.5 m but not closer to 1
m from the walls [1, 10].
Embedded earth electrodes shall be installed in such a way as to allow inspection during
construction. The embedded depth and type of the earth electrodes shall be such as to
minimize the effects of corrosion, soil drying and freezing and thereby stabilize the
equivalent earth resistance [10, 12]. It is recommended that the first metre of vertical
earth electrode should not be regarded as being effective under frost conditions.
The earth electrode design may be influenced by the soil conditions. The treatment of soil
resistivity can have corrosive effects on the earth electrode. The relationship between the
soil resistivity and corrosiveness is given in table 7-3 [12].
33
Table 7-3:
Soil resistivity and corrosiveness
Corrosiveness
Soil resistivity [Ω.m]
very severe
0 – 10
moderate to severe
10 – 100
Mild
100 – 1000
probably not corrosive
> 1000
34
8
Lightning protection system for thatched roofs
The discussion in chapters 6 and 7 gives some guidance on the lightning protection
system for a thatched roofed structure. There are two basic approaches to providing
sufficient protection for thatched roofs, that of lightning masts or a meshed conductor
system on the roof.
8.1 Lightning mast
According to SABS 0313, the lightning mast must be located at least 1 metre away from
the house, with sufficient height to provide an effective cone of protection. For small
structures one mast should be sufficient to form the cone of protection as shown in Figure
8-1. In the case of large structures and structures involving a group of closely spaced
houses, a minimum of two masts are recommended, with four being even more effective.
Figure 8-2 shows the implementation use of the two masts. The height of the mast and its
relative angles of protection are given on Table 6-1.
According to the estimated efficiencies in section 5.1.4, the lightning protection system
for thatched roofed structures is considered in this work taking into account the risk
levels (levels of protection) I and II. According to Table 6-1, the protective angles for
these levels of protection are 250 and 350 as shown in Figure 8-1.
35
Figure 8-1:
Lightning protection system using one mast
Figure 8-2:
Lightning protection system using two masts
36
In order to provide effective protection for thatched roofs, the zone of protection provided
by the mast must include gables ends, chimneys, antennas, vent pipes and any other metal
objects. Telephone wires, overhead service connections to the power system, or other
overhead metal wires or pipes must not enter the structure through or close to the thatch.
The power system and communication system must enter the building from bellow
ground level.
8.2 Lightning conductors above the roof
Protecting thatched roofs against lightning using a conductor system above the roof is
considered in this work as an alternative method for the method described in paragraph
8.1 above. A possibly more effective protection can be expected from a lattice of
lightning conductors strung some distance above the roof. As described in section 6.3.
Since the thatch is prone to ignite, the side-flash becomes a problem. This should be
avoided by considering the recommended size of the mesh on the roof given on Table 6-3
and the down conductor for the mesh conductors have to be routed far enough away from
the thatched roof
8.3 Protection of equipment inside the thatched roofed structure
In order to complete a lightning protection system design, potential equalization is a
fundamental basis for the realization of internal lightning protection, that is the lightning
over-voltage protection for the electrical and also the electronic data transmission
facilities and devices in a thatched roofed structure. In the event of a lightning stroke, the
potential of all installations in the affected building (including live conductors in the
electrical systems with arrestors) will be increased to a value equivalent to that arising in
the earthing system and no dangerous over-voltages will be generated in the system.
37
The protection of electronic equipment against potential differences and static charge
build up will be provided by interconnecting all non-current carrying metal objects to a
bonding bar that is effectively connected to the earth electrode system.
38
9
High voltage laboratory tests
The tests aim to verify the resistibility of thatch to ignite when hit by lightning. These
tests provide a mechanism to determine the behaviour of thatch when side-flash occurs
and from this appropriate protection measures can be implemented
The lightning current test is performed according to SABS IEC 1312-1:1995 and SABS
IEC 60-1:1989 standards. These standards specify the tolerance parameters for the
different exponential current impulses as shown on Table 9-1 and give a guideline on
how a lightning current impulse is obtained for test purposes. The tolerances defined
below are for the testing of surge protective devices and as no test standard is available
for thatch these parameters will be adopted for this work.
Table 9-1:
Standard exponential impulse current according to SABS IEC 60-1
Waveform
Time to crest
Time to half value
Value for peak
1/20
1µs ±10%
20 µs ±10%
±10%
4/10
4µs ±10%
10 µs ±10%
±10%
8/20
8µs ±10%
20 µs ±10%
±10%
30/80
30µs ±10%
80 µs ±10%
±10%
9.1 Test description
A combined impulse generator (voltage, current) and associated measuring instruments
were used for the testing of old thatch and wet thatch.
Figure 13-1 in appendices shows the test and measuring system which comprises a
combined impulse generator (voltage, current) ranging from 0 − 40 kA and measuring
instrument (TEKTRONIX THS 720P). With prefixed air-gap length of 10 mm, the
39
voltage was varied through the voltage control unit until the air broke down, and then the
peak current was measured. These tests are accompanied by a spark and some noise
which indicate the occurrence of a flash-over within the spark gap. The tests were
repeated for a fixed air-gap length of 20 mm and 30 mm.
The experimental test setup is shown in Figure 13-1 in appendices. The results of these
measurements are shown on Tables 9-2, 9-3 and 9-4 below. Figures 9-1, 9-2 and 9-3
show the impulse current waveforms.
9.2 Discussions of results
The test showed that the old thatch can easily ignite compared to the wet thatch. Some
parts of the old thatch started to smoke when the current was increased up to 10 kA. At
32 kA the noise and the spark became more intense. Because of safety issues, the test
with 32 kA to 40 kA was made without thatch in order to avoid fire since the high voltage
laboratory at the School of Electrical and Information Engineering in the University of
the Witwatersrand, Johannesburg is not designed for fire tests. The wet thatch was
subjected to the same conditions as the old thatch, it did not burn but the area affected by
the spark dried up.
During the test, it was also observed that when the airgap is small the spark and the noise
are more intense than when the airgap is big. With an air-gap of 10 mm the thatch started
to smoke at 10 kA with related impulse charge of 20 C. With an air-gap of 20 mm and 30
mm a current higher than 10 kA would be needed in order for the thatch to ignite. This
can be seen in the result measurements on Tables 9-2, 9-3 and 9-4
Four peak current tests using an air-gap of 10mm were carried out. The four peak current
tests and their respective results are shown on Table 9-2 and Figure 9-1 below.
40
Table 9-2:
Parameters from test with 10 mm of airgap
Parameters
Measured parameters
1
2
3
4
5
8
10
20
Peak rate-of-rise [kA/µs]
0.58
0.69
1.2
3.5
Time to crest [µs]
1.1
1.3
1.6
3.2
Time to half value [µs]
2.3
2.8
3.2
7.5
Impulse charge [C]
6.5
12
20
45
Sparking
No
No
Yes
Yes
Smoking
No
No
Yes
Yes
Ignition
No
No
No
Yes
Peak current [kA]
Figure 9-1:
Impulse current waveforms from measurements with 10 mm of
airgap
41
The measured impulse current waveforms are within the tolerance defined in Table 9-1,
therefore these results are acceptable.
With air-gaps of 20 mm and 30 mm, four peak current tests were carried out for each airgap. The results of the tests are presented on Tables 9-3 and 9-4 below. Figures 9-2 and
9-3 show the impulse current waveforms. As well as with tests performed at a gap length
of 10 mm the new results are also within the acceptable tolerance as described in Table 91.
Table 9-3:
Parameters from test with 20 mm of airgap
Parameters
Measured parameters
1
2
3
4
Peak current [kA]
20
25
30
35
Peak rate-of-rise [kA/µs]
1.2
3
4.5
5.6
Time to crest [µs]
9.8
11.6
14.5
21.6
Time to half value [µs]
18.5
32.5
46.7
49.8
Impulse charge [C]
80
110
160
186
Sparking
Yes
Yes
Yes
Yes
Smoking
Yes
Yes
Yes
x
Ignition
Yes
Yes
Yes
x
x – the test was done without thatch
42
Figure 9-2:
airgap
Impulse current waveforms from measurements with 20 mm of
Table 9-4:
Parameters from test with 30 mm of airgap
Parameters
Measured Parameters
1
2
3
4
Peak current [kA]
25
30
36
39
Peak rate-of-rise [kA/µs]
1.5
2.5
3.2
4.3
Time to crest [µs]
24.7
25.1
43
46.1
Time to half value [µs]
75.5
83.4
98.1
120
Impulse charge [C]
86
109
150
287
Sparking
N
Yes
Yes
Yes
Smoking
N
Yes
x
x
Ignition
N
N
x
x
x – the test was done without thatch
43
Figure 9-3:
Impulse current waveforms from measurement with 30 mm of airgap
The measured parameters such as peak current amplitude, peak rate-of-rise, and time
duration and impulse charge shown on Tables 9-2, 9-3 and 9-4 and waveforms in Figures
9-1, 9-2 and 9-3 are of primary importance for evaluation of the direct lightning effect in
thatch. These parameters are to be applied for thatched roofed structures lightning
protection design.
The waveforms in Figure 9-1, 9-2, and 9-3 are not intended to replicate a specific
lightning event, but they are intended to be composite waveforms whose effects on thatch
are those expected from natural lightning.
44
10 Conclusions and recommendations
From the literature reviews and the test results that culminate with the present research
report , it is possible to cite the following conclusions and recommendations:
The results obtained from the various tests show that a lightning current of 10kA with an
impulse charge of 20 C and with the waveform 1 in Figure 9-1 can be sufficient to cause
the thatch to smoke. Therefore with a lightning current of 10kA, the probability of the
occurrence of side-flashes is high when there are metal objects close to the thatched roof.
The thatch started to ignite at peak current of 20 kA and the impulse charge of 45 C with
impulse current waveform 4 in Figure 9-1.
The lightning protection system for thatched roofed structures should be designed by a
qualified technician according to standards.
There is a limited amount of information on lightning protection of thatched roofed
structures. This work contributes to this pool of knowledge and will provide useful
information for further research.
From the test results, it is clear that thatch can ignite if side-flash occurs between metal
objects. This can be avoided if the recommended clearance distances between thatch and
all lightning conductor objects are considered. These conductors must be bonded to the
earth termination system of the lightning protection system.
One of the main aims of this work was to test the performance of lightning protection
system for thatched roofed structures, The test facilities used however is not designed for
fire tests, therefore a complete structure test was not possible. Therefore only the thatch
resistibility tests were performed.
45
From calculation of the efficiency of LPS in section 5.1.4, the protective levels that can
be used for the LPS of thatched roofed structures are I and II
46
11 List of references
[1]
SABS 0313 (1999): “The protection of structures against lightning” SABS, 1999.
[2]
Geraldo Kindermann: “Proteção Contra Descargas Atmosféricas em Estruturas
Edificadas 3a Edicão, Brasil, 2003.
[3]
Williams ER: “The electrification of thunderstorms” Scientific American,
November 1988. pp 48-65.
[4]
William C. Hart and Edgar W. Malone: “Lightning and Lightning Protection”,
USA, 1979.
[5]
KJ Nixon, JM Van Coller and IR Jandrell: “ELEN 5011-Earthing and Lightning
protection”, Lecture notes, University of The Witwatersrand South Africa, 2005.
[6]
G. A. M. Odam: “Effects of lightning on asset, facility or structure”, National
Lightning Safety Institute (NLSI),
http://www.lightningsafety.com/.nlsi_lhm/effect.html, Last Accessed 28 January
2006.
[7]
SABS IEC 61024-1-1 (1993): “Protection of structures against lightning: Part 1:
General principles: Section 1: Guide A –Selection of protection levels for
lightning protection systems”, IEC, 1993
[8]
SABS IEC 61662 (1995): “Assessment of risk of damage due to lightning” IEC,
1995.
[9]
IEC 62305-2 Ed.1: “Protection against lightning – Part 2: risk management”,
2004.
[11]
BS6651. “British Standard code the protection of structures against lightning”,
BS, 1985
[12]
SABS 10199 (1985): “The design and installation of an earth electrode” SABS,
1985.
[14]
SABS IEC61312-1 (1995): “Protection against lightning electromagnetic impulse
47
12 Appendices
12.1 Lightning ground flash density, Ng
Table 12-1:
Lightning ground flash density Ng in South Africa [1]
48
12.2 Determination of the environmental coefficient, Ce
Table 12-2:
Environmental coefficient, Ce [1]
12.3 Calculation of the accepted annual frequency of lightning flashes,
Nc
Table 12-3:
Factors depending on the type of structure [1]
Description
Factor
Construction of walls
a1
Reinforced concrete interconnected metal façade
5
Interconnected prefabricated reinforced concrete parts, steel skeleton
4
Masonry, concrete without steel, not interconnected prefabricated
0.5
concrete parts
Wood framework or other combustible materials
49
0.1
Roof construction
a2
Steel
4
Reinforced concrete
2
Prefabricated reinforced concrete parts
0.5
Wood
0.1
Roof covering
a3
Reinforced concrete
4
Metal sheet
2
Tile, slate
1
plastic foils, roofing felt
0.5
Thatched roofs
0.05
Equipment on the roof
a4
No equipment
1
No earthed equipment, antennas
0.5
Electric apparatus
0.2
Sensitive electric parts
0.1
a = a1 × a 2 × a 3 × a 4
a1 = 0,5 ; a 2 = 0,1 ; a 3 = 0,05 ; a 4 = 1
Table 12-4:
Factors depending on the contents of structure [1]
Description
Factor
Risk of panic
b1
No risk of panic
1
Normal risk of panic
0.1
High risk of panic
0.01
Kind of contents
50
b2
Non combustible
1
Combustible
0.2
Explosive
0.1
Power mill
0.01
Nuclear plant
0.001
Value of contents
b3
Valuable
1
Very valuable
0.1
Irreplaceable
0.01
Measures for reduction of damage
b4
Automatic fire extinguishing system
10
Fire resisting structures
5
Fire alarm system
2
No measures
1
b = b1 × b2 × b3 × b4
b1 = 1 ; b2 = 0,2 ; b3 = 1 ; b4 = 2
Table 12-5:
Factors depending on the consequential losses [1]
Description
Factor
Danger to the environment
c1
No
1
Common
0.5
High
0.1
Very high
0.01
Loss of services to the public
No
c2
1
51
High
0.1
Very high
0.01
Other consequential losses
Low
c3
1
Normal
0.5
High
0.1
Very high
0.01
c = c1 × c 2 × c3
c1 = 1 ; c2 = 1 ; c3 = 1
12.4 Test and measurement system
Figure 12-1: Test and measurement system
52
12.5 Equivalent circuit
Figure 12-2: Equivalent circuit of current impulse generator (8/20 µs) [13]
Figure 12-3: Equivalent circuit of combined (Voltage, Current) impulse generator
(1.2/50 µs, 8/20 µs) [13]
53